Optimizing Simulation Efficiency in the LANL Neutron EDM Experiment Using Trajectory-Based Filtering

The Los Alamos National Laboratory Neutron Electric Dipole Moment (LANL nEDM) experiment aims to measure the neutron’s electric dipole moment using ultracold neutrons and Ramsey’s method of separated oscillatory fields. As the only nEDM experiment currently active in the United States, its success relies heavily on a robust simulation pipeline to guide design, optimize performance, and standardize data analysis. The goal of this summer research project was to restructure the simulation framework into a more efficient system that could yield improved statistics without increasing computational cost. To achieve this, we developed software capable of analyzing simulation output and implemented a strategy to divide the simulation into two distinct phases. The first phase tracks particle behavior up to a predefined boundary, allowing us to identify which particles are unlikely to reach the detector. By analyzing these trajectories, we aimed to establish a predictive trend that could replace the first phase entirely—eliminating unnecessary computation on non-contributing particles. Although we successfully configured the first simulation stage and its boundaries, the limited dataset prevented us from generating a reliable trend. As a result, the optimization remains incomplete and will require further data collection to validate the approach.